System to detect underground objects using a sensor array

Abstract

Systems and method to detect presence of buried landmines in a suspect area. A drone is outfitted with a ground penetrating radar, an infrared camera, and a metal detector mounted onto a leveling platform. The drone is flown over the suspect area while maintaining the leveling platform horizontal. Signals from the ground penetrating radar, an infrared camera, and a metal detector are converted into radargram, thermal image, and metal gram. Convolutional neural networks are applied to each of the into radargram, thermal image, and metal gram to detect anomalies.

The invention claimed is: 
     
       1. A system for detecting buried items, comprising:
 an airborne platform; 
 an actively leveling platform coupled to the airborne platform and operable to actively maintain the leveling platform in a horizontal orientation; 
 a ground penetrating radar (GPR) mounted on the actively leveling platform; 
 a metal detector mounted on the actively leveling platform; 
 an infrared (IR) camera mounted on the actively leveling platform; 
 wherein the metal detector comprises a very low frequency (VLF) coil arrangement and a pulse induction coil. 
 
     
     
       2. The system of  claim 1 , wherein the airborne platform comprises a drone, and the actively leveling platform is mounted below the drone. 
     
     
       3. The system of  claim 2 , wherein the drone comprises a plurality of motorized spools operable to dispense and retrieve cables, and wherein one end of each of the cables is attached to the actively leveling platform. 
     
     
       4. The system of  claim 2 , wherein the drone comprises a main body in the shape of an isotoxal star and having four arms, a plurality of stalks, each pair of stalks connected to one of the four arms at one end and supporting a plurality of motors at opposite end, the stalks and the arms defining a four-square grid with the main body located centrally to the four-square grid. 
     
     
       5. The system of  claim 4 , wherein each of the four arms includes a leveling arrangement to maintain the leveling platform horizontally during flight. 
     
     
       6. The system of  claim 2 , wherein the drone comprises a plurality of motorized actuators, each operable to extend and retract an extension rod, and wherein one end of the extendible rod is attached to the actively leveling platform. 
     
     
       7. The system of  claim 1 , wherein the ground penetrating radar operates in two frequencies, one selected in MHz range and one selected in GHz range. 
     
     
       8. The system of  claim 1 , wherein the VLF coil arrangement comprises a transmit and a receive coil arranged in a double-D topography. 
     
     
       9. The system of  claim 1 , further comprising a signal processing module receiving signals from the GPR, IR camera, and metal detector and generating a radargram, a thermographic map, and a metal detection map. 
     
     
       10. The system of  claim 9 , wherein the signal processing module further generates a tomographic map. 
     
     
       11. The system of  claim 9 , wherein the signal processing module is configured to perform the process comprising,
 scanning over a suspect area radar energy waves into the earth from the GPR and receiving a reflected radar signals; 
 scanning the suspect area with the infrared (IR) camera to generate IR thermal signals; 
 scanning the suspect area with a magnetic field from the metal detector and detecting changes in magnetic flux to generate magnetic flux signals; 
 generating the radargram from the reflected radar signals; 
 generating the thermographic map from the IR thermal signals; 
 generating a metal detection map from the magnetic flux signals; 
 applying convolutional neural network to each of the radargram, the thermographic map and the metal detection map to assign a probability score to each of the radargram, the thermographic map and the metal detection map, each of the probability scores indicating a probability that an item is buried in the suspect area; and 
 generating an overall probability score using the probability scores assigned to the radargram, the thermographic map and the metal detection map. 
 
     
     
       12. The system of  claim 11 , wherein scanning the suspect area comprises flying the airborne platform over the suspect area. 
     
     
       13. The system of  claim 12 , further comprising maintaining a leveling platform of the airborne platform horizontal during the scanning of the suspect area. 
     
     
       14. A system for detecting buried items comprising:
 an airborne platform; 
 an actively leveling platform coupled to the airborne platform and operable to actively maintain the leveling platform in a horizontal orientation; 
 a ground penetrating radar (GPR) mounted on the actively leveling platform; 
 a metal detector mounted on the actively leveling platform; 
 an infrared (IR) camera mounted on the actively leveling platform; 
 and further comprising a processing module having three convolutional neural networks, each receiving a signal from one of the GPR, IR camera, and metal detector and generating a probability score indicating a probability that a buried object has been detected. 
 
     
     
       15. The system of  claim 14 , wherein the metal detector comprises a very low frequency (VLF) coil arrangement and a pulse induction coil. 
     
     
       16. The system of  claim 14 , wherein the ground penetrating radar operates in two frequencies, one selected in MHz range and one selected in GHz range. 
     
     
       17. The system of  claim 14 , wherein the signal processing module generates a radargram, a thermographic map, and a metal detection map. 
     
     
       18. The system of  claim 17 , wherein the signal processing module further generates a tomographic map. 
     
     
       19. The system of  claim 14 , wherein the airborne platform comprises a drone, and the actively leveling platform is mounted below the drone. 
     
     
       20. The system of  claim 14 , wherein the convolutional neural network assigns a probability based on comparison to a threshold. 
     
     
       21. The system of  claim 14 , wherein the convolutional neural network assigns a probability based on comparison to a known anomaly signature. 
     
     
       22. The system of  claim 14 , wherein the convolutional neural network generates an overall probability score by applying weights to the probability score of the radargram, the thermographic map and the metal detection map. 
     
     
       23. The system of  claim 22 , wherein the weights are assigned according to particular consideration, comprising:
 a. Whenever (or when only) the GPR detects an anomaly sizable enough to be a potential mine, it is flagged and given the minimum probability rating; 
 b. Whenever the GPR and either the IR camera or the metal detector detects anomalies at the same location, it is flagged and given at least a minimum landmine probability rating; 
 c. Whenever only metal detection and GPR detect anomalies, Infrared score is disregarded; 
 d. Whenever Infrared and GPR ratings are above the landmine threshold and metal is not detected, metal detection score is disregarded; 
 e. Whenever Infrared and GPR ratings above threshold and metal is low but not negligible, score is increased; 
 f. Whenever either Infrared or Metal Detection detect anomalies where GPR does not, the system is paused and reviewed for false negatives; and 
 g. Whenever both Infrared and Metal Detection detect anomalies above threshold where GPR does not, the system identifies a detection error and the field survey is stopped. 
 
     
     
       24. A system for detecting buried items, comprising:
 an airborne platform; 
 an actively leveling platform coupled to the airborne platform and operable to actively maintain the leveling platform in a horizontal orientation; 
 a ground penetrating radar (GPR) mounted on the actively leveling platform; 
 a metal detector mounted on the actively leveling platform; 
 an infrared (IR) camera mounted on the actively leveling platform; 
 a signal processing module receiving signals from the GPR, the IR camera, and the metal detector and generating a radargram, a thermographic map, and a metal detection map; 
 wherein the signal processing module is configured to perform the process comprising: 
 receiving a reflected radar signal corresponding to a scanned area from the GPR; 
 receiving infrared (IR) signals corresponding to the scanned area from the IR camera; 
 receiving magnetic flux signals corresponding to the scanned area from the metal detector; 
 generating the radargram from the reflected radar signals; 
 generating the thermographic map from the IR signals; 
 generating the metal detection map from the magnetic flux signals; 
 applying convolutional neural network to each of the radargram, thermographic map and metal detection map to assign a probability score to each of the radargram, thermographic map and metal detection map, each of the probability scores indicating a probability that an item is buried in the suspect area; and 
 generating an overall probability score using the probability scores assigned to the radargram, the thermographic map and the metal detection map. 
 
     
     
       25. The system of  claim 24 , wherein the convolutional neural network assigns a probability based on comparison to a threshold. 
     
     
       26. The system of  claim 24 , wherein the convolutional neural network assigns a probability based on comparison to a known anomaly signature. 
     
     
       27. The system of  claim 24 , wherein generating an overall probability score comprises applying weights to the probability score of the radargram, the thermographic map and the metal detection map. 
     
     
       28. The system of  claim 27 , wherein the weights are assigned according to particular consideration, comprising:
 a. Whenever (or when only) the GPR detects an anomaly sizable enough to be a potential mine, it is flagged and given the minimum probability rating; 
 b. Whenever the GPR and either the IR camera or the metal detector detects anomalies at the same location, it is flagged and given at least a minimum landmine probability rating; 
 c. Whenever only metal detection and GPR detect anomalies, Infrared score is disregarded; 
 d. Whenever Infrared and GPR ratings are above the landmine threshold and metal is not detected, metal detection score is disregarded; 
 e. Whenever Infrared and GPR ratings above threshold and metal is low but not negligible, score is increased; 
 f. Whenever either Infrared or Metal Detection detect anomalies where GPR does not, the system is paused and reviewed for false negatives; and 
 g. Whenever both Infrared and Metal Detection detect anomalies above threshold where GPR does not, the system identifies a detection error and the field survey is stopped.